problme with aggregator stage

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
supernova2005
Participant
Posts: 37
Joined: Mon Jan 24, 2005 10:12 am

problme with aggregator stage

Post by supernova2005 »

Hello everyone,

I have an aggregator stage that calculates the sum of a decimal(13,2) field. Both input and output are defined as not nullable. The result of is correct, but I am getting the following warning message:


Aggregator_66: When checking operator: When binding output interface field "value1" to field "value1": Converting nullable source to non-nullable result; fatal runtime error could occur (use modify operator to specify value to which null should be converted)

Does anyone know what causes this warning message and how I can get rid of it?

Thanks.
amsh76
Charter Member
Charter Member
Posts: 118
Joined: Wed Mar 10, 2004 10:58 pm

Post by amsh76 »

Whats your source?
supernova2005
Participant
Posts: 37
Joined: Mon Jan 24, 2005 10:12 am

Post by supernova2005 »

amsh76 wrote:Whats your source?
both the source and the target are sequential files. but I don't think it is a problem, because I tried to replace them with datasets and the same thing happened.
RobertScarbrough
Participant
Posts: 24
Joined: Fri Oct 01, 2004 1:43 pm
Location: USA

Re: problme with aggregator stage

Post by RobertScarbrough »

From what I have read the only way to get rid of the message is to catch the field in a modify stage or a transformer in the subsequent output stage from the aggregator. I have talked with some folks from Ascential about it and it seemed to be like that is just the way it is.
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Supernova,

You will need to put in a Modify stage where you cast the datatype, or a TRANsform stage that does the same and also puts in a derivation that ensures no NULLs are passed down that column.

Unlike server, where the NULLable attribute really only applies when reading/writing to a Database stage, in PX you need to ensure continuity of attributes.
ds_is_fun
Premium Member
Premium Member
Posts: 194
Joined: Fri Jan 07, 2005 12:00 pm

Post by ds_is_fun »

The way AGGR stage works on sum of decimal type columns is. When it performs a SUM it makes sure that no row is "null". When we contacted an Ascential pro' with the same problem. He mentioned it was a compiler related error, where the code needs to changed to avoid this problem in future.
I changed the datatypes to the condition nullable = 'Y' in the Output of the AGGR stage.
Then I threw an additional stage, a transformer(most of the Architects would'nt like this). nullable from Input obviously is a 'Y' and change the output condition of nullable to 'N'. In the trasformer have derivation defaulting as IfIsNull(coulmn_name) then <value>.
This should get rid of the problem.
Good luck!
supernova2005
Participant
Posts: 37
Joined: Mon Jan 24, 2005 10:12 am

Post by supernova2005 »

Yes, you are right. The warning messages disappeared. Thank you for the help.

ds_is_fun wrote:The way AGGR stage works on sum of decimal type columns is. When it performs a SUM it makes sure that no row is "null". When we contacted an Ascential pro' with the same problem. He mentioned it was a compiler related error, where the code needs to changed to avoid this problem in future.
I changed the datatypes to the condition nullable = 'Y' in the Output of the AGGR stage.
Then I threw an additional stage, a transformer(most of the Architects would'nt like this). nullable from Input obviously is a 'Y' and change the output condition of nullable to 'N'. In the trasformer have derivation defaulting as IfIsNull(coulmn_name) then <value>.
This should get rid of the problem.
Good luck!
Post Reply